COMPARISON BETWEEN EKTACHEM SLIDE TECHNOLOGY AND ATOMIC ABSORPTION SPECTROSCOPY METHOD FOR LITHIUM DETERMINATION IN SERUM.
Nelson Geraldino, Charles Williams, Mary Jane Horenzy, Mohamed Virji, Warren Diven and Vijay Warty (Department of Pathology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213)
Lithium is used for treatment of the manic phase of affective disorders, mania, and manic-depressive illness. Because of its sedative effect on the central nervous system, it is very important to measure the lithium concentration in serum or plasma to monitor patient compliance and to avoid possible overdose. Two assays using very different methodologies to measure the lithium levels in serum were compared. Our reference method is atomic absorption (AA) spectrophotometry. We evaluated a dry-slide method on Ektachem instrument that is based on the specific binding of lithium ion to a crown-ether chromophore conjugate. Serum specimens from 33 patients who were on lithium therapy were assayed by both methods (range = 0.29-1.55 mEq/L) in the initial part of our study. Correlation showed that the Ektachem method underestimates the lithium concentration ( Ektachem = 0.8966 AA - 0.046, r2 = 0.975 ). This could have clinical implications at the higher end of the therapeutic range, since a concentration greater than 1.5 mEq/L as measured by AA is potentially toxic, but the Ektachem results may still be within the therapeutic range. In the second part of our study, we analyzed 25 serum specimens by both methods, the Ektachem method being modified using new Supplementary Analyte Values (SAVs) (range = 0.10-1.46 mEq/L). The new correlation between the two methods did not indicate any bias (Ektachem = 0.9796 AA + 0.002, r2 = 0.994). This data demonstrates that the dry-slide assay using modified SAVs by Ektachem agrees very well with AA spectrophotometry method.
Introduction
Lithium in carbonate and citrate forms is used to treat manic-depressive illness. Lithium ions are almost completely absorbed from the gastrointestinal tract within eight hours after oral administration. The therapeutic range for lithium in the serum is 0.6 - 1.5 mEq/L, and patients have to be constantly monitored for levels of the drug because of potential toxicity, symptoms of which are tremors, nausea, vomiting, drowsiness, decreased coordination, and possible seizures and coma, all of which are quite nonspecific.
The method currently used in our laboratory to monitor serum lithium levels makes use of atomic absorption spectrophotometry, which is the reference assay method. However, the procedure is time consuming, with a turnaround time in hours, and technologists have to be specially trained to use the machine.
Johnson and Johnson has developed a dry-slide method utilizing reflectance spectrophotometry to determine lithium levels in the serum. Potential advantages of the procedure include a shorter turnaround time, less complicated specimen preparation, and less training involved in the operation of the Ektachem analyzer, which is currently being used in a number of other assays in the clinical chemistry laboratory.
Determination of lithium can likewise be done using flame photometry, advantages of which over the atomic absorption method are similar to those of the dry-slide method.
Objective and Methodology
The objective of this study is to compare the accuracy of the Johnson and Johnson dry-slide method for the determination of lithium levels in serum with the reference method, which is atomic absorption spectroscopy, prior to eventual switch to the former assay for routine use in the clinical chemistry laboratory.
In atomic absorption spectrophotometry, the element is dissociated from its chemical bonds and placed in an unexcited or ground state (neutral atom). This means that the neutral atom is at a low energy level in which it is capable of absorbing radiation at a very narrow bandwidth corresponding to its own line spectrum. A hollow cathode lamp with the cathode made of the material to be analyzed is used to produce a wavelength of light specific for the material. Thus, if the cathode were made of lithium, lithium light at predominantly 670.8 nm would be emitted by the lamp. When the light from the hollow cathode lamp enters the flame, some of it is absorbed by the ground-state atoms in the flame, resulting in a net decrease in the intensity of the beam from the lamp.
In the dry-slide technology, the sample is deposited on the slide and is evenly distributed by the spreading layer. It then traverses the buffer and barrier layers to rest on the dye layer, where lithium is specifically bound by the crown-ether chromophore conjugate(6-dodecyl-6-(2'-hydroxy-5'-(2",4"-dinitrophenylazo)benzyl)-13,13-dimethyl-1,4,8,11-tetraoxacyclotetradecane). As the lithium ion binds to the crown-ether, a shift in the peak absorbance of the chromophore conjugate occurs. The increase in absorbance at 600 nm is proportional to the concentration of lithium in the sample, and the intensity of the dye is measured by reflectance spectrophotometry at 2.3 minutes.
To do the initial comparison studies, we utilized patient specimens which had been previously analyzed by atomic absorption spectroscopy. Thirty-three patient specimens were assayed, with concentrations ranging from 0.29 mEq/L to 1.55 mEq/L.
Results
Regression analysis (33 patient samples with AA concentrations ranging from 0.29 to 1.55 mEq/L) was performed using Quattro Pro 5.0 (Borland). The squared correlation coefficient was 0.9747, with an x coefficient of 0.8966. This means that the proposed assay (dry-slide) was giving out results which were about 10 % lower than the reference assay (atomic absorption). The distribution of points along the regression line as well as a summary of the regression output is shown in Fig. 1.
The finding that the dry-slide assay gives results which were about 10 % lower than the reference method has important implications in the routine monitoring of the drug. Lithium is considered to be potentially toxic at levels beyond 1.5 mEq/L, and in that range levels which are already beyond the therapeutic range as determined by atomic absorption spectroscopy may still be within the normal range as determined by the proposed assay.
In December 1994 Johnson and Johnson came out with new SAVs (supplementary assay value) to address the problem on the inadequacy of the x coefficient and its possible clinical implications. This time we compared 25 patient samples which were previously analyzed for lithium levels on atomic absorption spectroscopy with two Ektachem analyzers (an Ektachem 250 and an Ektachem 700). These samples had lithium levels ranging from 0.1 to 1.46 mEq/L.
The regression output from this comparison study showed the square of the correlation coefficient for AA vs. the Ektachem to be 0.9941. The x coefficient for the same data set was 0.9796. Both of these parameters showed marked improvement with the introduction of new SAVs. The distribution of points along the regression line is shown in Fig. 2.
Conclusions
- Determination of lithium using dry-slide technology on Ektachem has generated results which were about 10 % lower than those given by the reference assay, which is atomic absorption spectroscopy. This could have important clinical implications especially at the higher end of the range, where the proposed procedure may be giving results which were still within the therapeutic range, but may actually be on the potentially toxic range when measured by the reference assay.
- With the introduction of new SAVs (supplementary assay value), the correlation studies match very well, giving justification for the change in assay methodology in the automated chemistry laboratory.
- The coefficient of variation at the low end of the dry-slide assay method is 4.66 %, and at the higher end it is 1.49 % (data not shown). Both are deemed acceptable for clinical use.
wwwadmin@path.upmc.edu
Who's who -- wwwadmin
